-
公开(公告)号：EP3879457A3
公开(公告)日：2022-01-12
申请号：EP21180631.0
申请日：2021-06-21
发明人： YANG, Fukui , WEN, Shengzhao , HAN, Junyu
IPC分类号： G06K9/62
摘要： The present disclosure provides a method, and an apparatus for model distillation, relates to the technical field of artificial intelligence, and in particular, relates to technical fields of deep learning and computer vision. A specific implementation includes: obtaining a batch of teacher features corresponding to a teacher model and a batch of student features corresponding to a student model; determining a set of teacher similarities corresponding to the batch of teacher features and a set of student similarities corresponding to the batch of student features; determining weights of loss values of features of images based on difference values corresponding to the images; and weighting a loss value of a feature of each image in a batch of images, training the student model by using a weighting result. The present disclosure may use the difference values between the feature similarities of the student model and the feature similarities of the teacher model to determine the weights of the loss values. The distillation process of the present disclosure may improve the detection capabilities of the models, reduce the delay of the execution devices, and reduce the occupation and consumption of computing resources such as memories.
-
公开(公告)号：EP3916634A3
公开(公告)日：2021-12-22
申请号：EP21180501.5
申请日：2021-06-21
发明人： ZHANG, Chengquan , LV, Pengyuan , YAO, Kun , HAN, Junyu , LIU, Jingtuo
摘要： A text recognition method includes: acquiring an image including text information, the text information including M characters, M being a positive integer greater than 1; performing text recognition on the image to acquire character information about the M characters; recognizing reading direction information about each character in accordance with the character information about the M characters, the reading direction information being used to indicate a next character corresponding to a current character in a semantic reading order; and ranking the M characters in accordance with the reading direction information about the M characters to acquire a text recognition result of the text information.
-
公开(公告)号：EP3879457A2
公开(公告)日：2021-09-15
申请号：EP21180631.0
申请日：2021-06-21
发明人： YANG, Fukui , WEN, Shengzhao , HAN, Junyu
IPC分类号： G06K9/62
摘要： The present disclosure provides a method, and an apparatus for model distillation, relates to the technical field of artificial intelligence, and in particular, relates to technical fields of deep learning and computer vision. A specific implementation includes: obtaining a batch of teacher features corresponding to a teacher model and a batch of student features corresponding to a student model; determining a set of teacher similarities corresponding to the batch of teacher features and a set of student similarities corresponding to the batch of student features; determining weights of loss values of features of images based on difference values corresponding to the images; and weighting a loss value of a feature of each image in a batch of images, training the student model by using a weighting result. The present disclosure may use the difference values between the feature similarities of the student model and the feature similarities of the teacher model to determine the weights of the loss values. The distillation process of the present disclosure may improve the detection capabilities of the models, reduce the delay of the execution devices, and reduce the occupation and consumption of computing resources such as memories.
-
4.发明公开
公开(公告)号：EP3968287A2
公开(公告)日：2022-03-16
申请号：EP22151884.8
申请日：2022-01-17
发明人： QIN, Xiameng , LI, Yulin , HUANG, Ju , XIE, Qunyi , ZHANG, Chengquan , YAO, Kun , LIU, Jingtuo , HAN, Junyu
IPC分类号： G06V30/41
摘要： Provided are a method and apparatus for extracting information about a negotiable instrument, an electronic device and a storage medium. The method includes inputting (S101) a to-be-recognized negotiable instrument into a pretrained deep learning network and obtaining a visual image corresponding to the to-be-recognized negotiable instrument through the deep learning network; matching (S102) the visual image corresponding to the to-be-recognized negotiable instrument with a visual image corresponding to each negotiable-instrument template in a preconstructed base template library; and in response to the visual image corresponding to the to-be-recognized negotiable instrument successfully matching a visual image corresponding to one negotiable-instrument template in the base template library, extracting (S103) structured information of the to-be-recognized negotiable instrument by using the negotiable-instrument template.
-
5.发明公开
公开(公告)号：EP3869404A3
公开(公告)日：2022-01-26
申请号：EP21183783.6
申请日：2021-07-05
发明人： ZHANG, Yanlong , PENG, Mian , YANG, Zuncheng , HAN, Junyu , LIU, Jingtuo
摘要： Embodiments of the present disclosure provide a vehicle loss assessment method executed by a mobile terminal, a device, a mobile terminal, a medium and a computer program product. The present disclosure relates to the field of artificial intelligence, and particularly relates to a computer vision and deep learning technology. The implementation solution includes: acquiring at least one input image; detecting vehicle identification information in the at least one input image; detecting vehicle damage information in the at least one input image; and determining a vehicle loss assessment result on the basis of the vehicle identification information and the vehicle damage information. By utilizing the method provided by the embodiments of the present disclosure, vehicle loss assessment may be executed offline at the mobile terminal without a need of sending the captured image to a cloud, so that the effects of high real-time performance, small network latency, saving of network service resources and saving of network bandwidth expenses in a loss assessment process may be achieved.
-
公开(公告)号：EP3882817A3
公开(公告)日：2022-01-05
申请号：EP21180801.9
申请日：2021-06-22
发明人： HUANG, Ju , XIE, Qunyi , LI, Yulin , QIN, Xiameng , YAO, Kun , HAN, Junyu
摘要： The present disclosure discloses a method, apparatus and device for recognizing a bill, and a storage medium. The method comprises: acquiring a bill image; inputting the bill image into a feature extraction network layer of a pre-trained bill recognition model, to obtain a bill key field feature map and a bill key field value feature map of the bill image; inputting the bill key field feature map into a first head network layer of the bill recognition model, to obtain a bill key field; processing the bill key field value feature map by a second head network layer of the bill recognition model, to obtain a bill key field value, the feature extraction network layer being respectively connected with the first head network layer and the second head network layer; and generating structured information of the bill image based on the bill key field and the bill key field value.
-
公开(公告)号：EP3842960A3
公开(公告)日：2021-11-17
申请号：EP21170920.9
申请日：2021-04-28
发明人： NI, Zihan , SUN, Yipeng , YAO, Kun , HAN, Junyu , DING, Errui , LIU, Jingtuo , WANG, Haifeng
IPC分类号： G06F16/33 , G06F16/583 , G06K9/00 , G06F16/35 , G06K9/62
摘要： The disclosure provides a method for processing information, belonging to a field of artificial intelligence including computer vision, deep learning, and natural language processing. In the method, the computing device (120) recognizes (210) multiple text items (115) in the image (110). The computing device (120) classifies (220) multiple text items (115) into a first set (117) of name text items and a second set (119) of content text items based on semantics of the text items. The computing device (120) performs (230) a matching operation (125) between the first set (117) and the second set (119) based on a layout of the text items in the image, and determines matched name-content text items (130). The matched name-content text items (130) include a name text item and a content text item matching the name text item. The computing device outputs (240) the matched name-content text items (130).
-
公开(公告)号：EP3882817A2
公开(公告)日：2021-09-22
申请号：EP21180801.9
申请日：2021-06-22
发明人： HUANG, Ju , XIE, Qunyi , LI, Yulin , QIN, Xiameng , YAO, Kun , HAN, Junyu
摘要： The present disclosure discloses a method, apparatus and device for recognizing a bill, and a storage medium. The method comprises: acquiring a bill image; inputting the bill image into a feature extraction network layer of a pre-trained bill recognition model, to obtain a bill key field feature map and a bill key field value feature map of the bill image; inputting the bill key field feature map into a first head network layer of the bill recognition model, to obtain a bill key field; processing the bill key field value feature map by a second head network layer of the bill recognition model, to obtain a bill key field value, the feature extraction network layer being respectively connected with the first head network layer and the second head network layer; and generating structured information of the bill image based on the bill key field and the bill key field value.
-
公开(公告)号：EP3869398A2
公开(公告)日：2021-08-25
申请号：EP21180877.9
申请日：2021-06-22
发明人： ZHANG, Chengquan , EN, Mengyi , HUANG, Ju , XIE, Qunyi , QIN, Xiameng , YAO, Kun , HAN, Junyu , LIU, Jingtuo , DING, Errui
摘要： A method and apparatus for processing an image, a device and a storage medium. An implementation of the method includes: acquiring a template image, the template image including at least one region of interest; determining a first feature map corresponding to each region of interest in the template image; acquiring a target image; determining a second feature map of the target image; and determining at least one region of interest in the target image according to the first feature map and the second feature map.
-
公开(公告)号：EP3859605A2
公开(公告)日：2021-08-04
申请号：EP21163966.1
申请日：2021-03-22
发明人： GUO, Zhizhi , SUN, Yipeng , LIU, Jingtuo , HAN, Junyu
IPC分类号： G06K9/00
摘要： The present application discloses an image recognition method, apparatus, device, and a computer storage medium, which is related to a technical field of artificial intelligence, and in particular, to a technical field of image processing. The method includes: performing organ recognition on a human face image and marking positions of the human facial five sense organs in the human face image, obtaining a marked human face image; inputting the marked human face image into a backbone network model and performing feature extraction, obtaining defect features of the marked human face image outputted by different convolutional neural network levels of the backbone network model; and fusing the defect features of different levels that are located in a same area of the human face image, obtaining a defect recognition result of the human face image. Embodiments of the present application can improve the accuracy and efficiency of human facial defect recognition.
-
-
-
-
-
-
-
-
-
搜索结果
27 条记录 (耗时 0.017 秒)
